v0.7.3

Changed

• Signals to input to mapping classes are now normalised before fitting, then M0 rescaled back to the original scale. This
  means M0 bounds and initialisation are more consistent/appropriate for data from different vendors #226
• The limits of efficiency have been widened for MOLLI T1 mapping #229
• Upgrade codecov action to v4

v0.7.2

Changed

• T1 mapping data is now assumed to have been magnitude corrected if the first percentile of the first inversion time is
  negative rather than the minimum value. This should make it more robust to noise/preprocessing artefacts. #222 #223

Added

• Export the fit signals from mapping functions e.g. the expected T1 recovery at times TI given the fit values of T1 and M0. #221
• Warnings if there aren't many negative values in T1 mapping data. #222 #223
• get_fit_signal methods have been added to most mapping classes to allow the user to get the fit signal at a given time point.

Fixed

• Issue where resources sub-module was not included on PyPI

v0.7.2-rc.1

Added

• Export the fit signals from mapping functions e.g. the expected T1 recovery at times TI given the fit values of T1 and M0. #221

Fixed

• Issue where resources sub-module was not included on PyPI

v0.7.1

Changed

• ukat is now tested on MacOS #219

Added

• B1 rescaling as function and default output of StimFit #218

Fixed

• DWI mask export bug #217

v0.7.0

Changed

• ukat is now tested against Python 3.11
• Dependencies are now a little less strict for some packages

Added

• T2StimFit - A method of accounting for stimulated echoes when performing T2 mapping #207, #209
• R-Squared values for curve fitting maps #198 #205
• New PR template for releases #205 #210

Fixed

• Mapping should now scale better over large images/multiple cores #165 #205
• Quite a lot of PEP8 formatting issues

v0.6.5

Changed

• Lots of dependencies have been updated to their latest versions #203
• Dependencies are now set to be compatible versions (~=) rather than greater than a specific version (>=) #203

Added

• Number of downloads badge to readme

Fixed

• MTR outputs are now squeezed to remove bonus dimensions #202

v0.6.4

Changed

• Increased upper bounds of M0 in T1, T2 and T2* exponential fit. #196 #199
• B0 mask now keeps True voxels rather than False voxels, making it align with the rest of UKAT. #194 #195

Added

• Python 3.10 is now supported.

Fixed

• Fixed bug in B0 offset calculation. #200

v0.6.3

Added

• iSNR maps #192

v0.6.2

Added

• Rescaling of GE B1 maps. #190

Fixed

• Hopefully reference manual now works, built using Python 3.9. #189

v0.6.1

Changed

• Python 3.7 is no longer supported.

Added

• Whole kidney segmentation can now be performed using custom models. #187